Robert J. Matz and other employees of Hamilton Investments, Inc. lost their jobs and the non-vested portions of their retirement benefit plan in 1994 when the company was sold. Matz filed suit on behalf of himself and other terminated employees who were participants in the ERISA pension plan, claiming entitlement to the benefits as a result of a partial termination of the plan. To prove partial termination, he seeks to count both vested and non-vested participants and, because he believes that the partial termination was the result of a multi-year corporate reorganization, to aggregate terminations that occurred in multiple plan years. The District Court ruled that he could do both, and certified the issues to us for interlocutory appeal. We affirm.

In 1994, Household International, Inc. was the parent corporation of a varied group of companies. Its portfolio included mortgage companies, banking institutions, retail securities brokerages, insurance businesses and credit card companies. In August, 1994, Household began selling off some of its subsidiaries, beginning with Hamilton Investments, Inc., the company for which Robert Matz worked from March 28, 1989 until his termination on September 1, 1994.

Hamilton, through its parent corporation Household, provided a retirement benefit plan ("the Plan") to its employees. It was an employee benefit plan within the meaning of 29 U.S.C. sec.1002(2)(A). The Plan allowed participants to make payroll contributions which were matched by contributions from Hamilton. Although the participant's contributions vested immediately, Hamilton's contributions were subject to deferred vesting. Section 14.1 of the Plan provided that a participant became vested in Hamilton's contributions at a rate of 20% per year. Thus, a participant had to remain with Hamilton for five years before he became fully vested. If his service to the company ended before that time, he forfeited the non-vested portion of Hamilton's contribution.

When Matz's job ended he had a 60% vested benefit in his employer's matching contribution. He elected to take a distribution and was paid $27,914.10, which represented 100% of his contributions to the Plan and 60% of Hamilton's contributions. The remaining 40%, $7,289.92, was forfeited and used by Household to reduce its matching contributions.

Matz sues to recover his forfeited amount. If he can prove that there was a termination or partial termination of the Plan,1 ERISA affords him relief. 26 U.S.C. sec.411(d)(3) provides

A trust shall not constitute a qualified trust under sec.401(a) unless the plan of which such trust is a part provides that--

A. Upon its termination or partial termination . . . the rights of all affected employees to benefits accrued to the date of such termination, partial termination or discontinuance, to the extent funded as of such date, or the amounts credited to the employee's accounts are non- forfeitable.

ERISA does not, however, define what constitutes a partial termination. Treasury Regulation sec.1.411(d)-2(b)(1) provides some guidance

Whether or not a partial termination of a qualified plan occurs (and the time of such event) shall be determined by the Commissioner with regard to all the facts and circumstances in a particular case. Such facts and circumstances include the exclusion, by reason of a plan amendment or severance by the employer, of a group of employees who have previously been covered by the plan; and plan amendments which adversely affect the rights of employees to vest in benefits under the plan.

(emphasis added).

Matz alleges that a partial termination of the Plan occurred beginning in August, 1994 (with the sale of Hamilton) and ending in May, 1996. During that period, Household discontinued or sold Household Mortgage Services, Inc., various branches of Household Bank, F.S.B., and Alexander Hamilton Life Insurance Company. Matz believes that this series of corporate transactions were part of a single reorganization plan and resulted in the Plan's partial termination. For that reason, he seeks to combine all of the terminations by Household in all of those years. He also seeks to count all fully vested employee terminations as well as non-vested employee terminations.

The Plan, by contrast, contends that a partial termination analysis should look at individual plan years, not aggregated years. Furthermore, it argues that aggregation is inappropriate because there was no corporate reorganization that would justify such an approach. It also asks that only non-vested participants be counted. Finally, the Plan contends that any partial termination analysis must end on September 30, 1995, the date it issued a plan amendment vesting, fully and immediately, all participants.2 Pointing to the language of 26 U.S.C. sec.411(d)(3), the Plan states that participants who separated from service after September 30, 1995 cannot be "affected employees" because all Plan accounts became non-forfeitable at that time.

The parties briefed the issues to the District Court. The court, after careful analysis, ruled that both vested and non-vested participants could be counted to determine whether a partial termination occurred and that Matz could combine all terminations for 1994 through 1996. Matz v. Household International Tax Reduction Investment Plan, 1998 WL 851491 (N.D.Ill. 1998); Matz v. Household International Tax Reduction Investment Plan, 1999 WL 754659 (N.D.Ill. 1999). Upon the Plan's motion, the rulings were certified for interlocutory appeal. Now before us are the questions: (1) whether, for purposes of determining if the Plan was "partially terminated," the court's analysis should include fully-vested employees terminated after the 1995 Plan amendment; and (2) whether the plaintiff's partial termination analysis may include terminations that take place over multiple (here, three) plan years. We affirm the District Court.

The question of whether a partial termination has occurred is a question of law subject to de novo review. Sage v. Automation, Inc. Pension Plan and Trust, 845 F.2d 885, 890 (10th Cir. 1988). Although many courts have addressed the issue of partial termination, few have addressed the questions of whether fully vested participants must be counted and whether the counting period is an individual plan year or the aggregation of multiple plan years. Indeed, they are issues of first impression to this court. We find that Matz has standing to raise these issues and address each one in turn.

A. Counting Of Vested And Non-Vested Participants

When interpreting congressional statutes, we look first at the plain language of the statute. See Reves v. Ernst & Young, 507 U.S. 170, 177 (1993). If the language is clear and unambiguous, we apply the statute so as to give effect to its plain meaning. United States v. Hayward, 6 F.3d 1241, 1245 (7th Cir. 1993). We cannot do that here because, although 26 U.S.C. sec.411(d)(3) refers to a partial termination, it does not define it. Nor does it provide us with the framework for an analysis. Thus, we must search the legislative history for instruction.

The legislative history provides little more guidance than that found in the statute. The House and Senate Reports state only that "[e]xamples of a partial termination might include, under certain circumstances, a large reduction in the work force, or a sizable reduction in benefits under the plan." H.R.Rep. No. 807, 93rd Cong., 2d Sess., reprinted in 1974 U.S.C.C.A.N. 4670, 4731; S.Rep. No, 383, 93rd Cong. 2d Sess., reprinted in 1974 U.S.C.C.A.N. 4890, 4935. Although this is somewhat more helpful, it still does not provide us with a clear standard by which to judge a particular case.

Our sister circuit struggled with the same dilemma before inviting an amicus brief from the IRS. See Weil v. Retirement Plan Administrative Committee, 933 F.2d 106, 109-10 (2nd Cir. 1991) ("Weil III"). The Weil III decision, finding that vested and non-vested participants must be counted when determining whether a partial termination occurred, was the culmination of several decisions by the District Court and the Second Circuit. Initially, the District Court ruled that a partial termination had not occurred. Weil, 577 F. Supp. 781 (S.D.N.Y. 1984). The Second Circuit reversed. Weil, 750 F.2d 10 (2d Cir. 1984) ("Weil I"). On remand, the District Court, counting both vested and non- vested participants, found that a partial termination had occurred. Weil, 1988 WL 64862 (S.D.N.Y. 1988). The Second Circuit reversed again, holding that only non-vested participants should be considered. Weil, 913 F.2d 1045, 1050- 51 (2nd Cir. 1990) ("Weil II").

On rehearing, the Weil III court had the benefit of the IRS' position of the issue. Giving "great weight" and deference to the IRS' view that both vested and non-vested participants should be counted, the Second Circuit vacated, in part, its ruling in Weil II and held that both classes of participants were to be considered in a partial termination analysis. Weil, 933 F.2d at 110. In so doing, the court found that the IRS' interpretation was reasonable and judicial deference was required as the IRS was the agency responsible for administering the partial termination statute. Id.

"Courts have generally held that termination of a number of employees does not constitute a 'partial termination' unless there is a significant reduction in plan participants." Matz, 1998 WL 851491, *2. To determine whether there is a significant reduction in plan participants, courts apply a "significant percentage" test. Kreis v. Charles O. Townley, M.D. & Associates, 833 F.2d 74, 79 (6th Cir. 1987). Partial termination is measured by using the ratio of terminated plan participants over total plan participants.

Matz urges us to include all terminated participants (vested and non-vested) in the numerator, while the Plan asks us to consider only the non-vested participants in the top of the equation. Both claim that their method best furthers the purposes of the statute. Once again, however, both the statute and its legislative history are silent as to the purpose of the partial termination statute. Courts, though, have generally understood the statute's purposes to be: (1) to protect employees' legitimate expectation of pension benefits; and (2) to prevent employers from abusing pension plans to reap tax benefits. Weil, 933 at 106, 107; Halliburton v. Commissioner of Internal Revenue, 83 T.C. 154, 161 (1984).

The Plan argues that the counting of vested participants would further neither of these purposes. It states "[t]he partial termination remedy is aimed at protecting employees from a forfeiture of unvested benefits, and preventing employers from receiving sizable, tax-free reversions of surplus plan assets." From a policy standpoint, the Plan makes a logical argument. Vested participants do not need further protection for their pension benefits and do not benefit from a finding of partial termination. Their benefits, by virtue of vesting, are non- forfeitable. The employer gains nothing either. No monies revert back to it because the benefits are vested and non-forfeitable. We agree with the Plan's arguments. And, as the District Court stated, "if [we] were writing on a blank slate, [we] would be inclined to consider only non- vested employees" in deciding whether a partial termination had occurred. However, we are not writing on a clean slate.

We are not alone in our view or our holding. The District Court for the Eastern District of Louisiana ruled that both vested and non-vested terminees must be considered, "[e]ven though the issue of partial termination affects only non- vested participants." Morales v. Pan American Life Insurance Company, 718 F. Supp. 1297, 1302 (E.D.La. 1989). Although the District Court's decision was appealed, the Fifth Circuit did not decide the issue, holding instead that plaintiff Morales, a vested employee, did not have standing to raise the issue on behalf of the non-vested employees. Morales v. Pan American life Insurance Company, 914 F.2d 83, 86 (5th Cir. 1990).

In a thorough and thoughtful opinion, the District Court for the Southern District of Texas also agreed that neither policy was furthered by the counting of vested participants. But, it ruled that only non-vested terminees would be counted. In re Gulf Pension Litigation, 764 F. Supp. 1149, 1165 (S.D.Tex. 1991). Realizing that approach would skew the percentage calculation and cause it to be artificially low (thereby increasing the chances of finding that there had not been a significant reduction in plan participants) the court excluded vested participants from both halves of the ratio. Id. at n.10. Having calculated in this manner, the court found a partial termination of the plan. On appeal, the Fifth Circuit did not consider whether a partial termination occurred and affirmed on other grounds. Borst v. Chevron Corporation, 36 F.3d 1308, 1314, n.11 (5th Cir. 1994) ("Because we do not consider whether or not a partial vertical (or horizontal) termination occurred, the district court's ruling on this issue is not conclusive between the parties.").

Purely from a policy standpoint, we believe that the method adopted in Gulf Pension Litigation best furthers the purposes of the partial termination statute. However, we are constrained in our analysis of the statute and must decide only whether the IRS' construction is reasonable. See Weil, 933 F.2d at 107-08. We conclude that it is. Neither the statute, its legislative history nor the Treasury Regulation mentioned above differentiate between vested and non-vested participants. Indeed, the Treasury Regulation says that a partial termination may occur when "a group of employees," previously covered by the plan, are excluded from the plan by reason of a severance from employment.

A finding of reasonableness is also supported by the fact that the exclusion of vested participants from the ratio calculation gives an inaccurate assessment of whether there has been a significant reduction in plan participation, the benchmark against which partial termination is measured. For the calculation to be accurate, the circumstances as a whole must be considered. This, we think, buttresses a finding that vested participants must be included.

"To uphold [the agency's interpretation] we need not find that [its] construction is the only reasonable one, or even that it is the result we would have reached had the question arisen in the first instance in judicial proceedings. . . We need only conclude that it is a reasonable interpretation of the relevant provisions." Weil III, 933 F.2d at 107-08 (internal quotation marks and citations omitted.). For the above reasons, we find that there is a factual basis for finding that the IRS' interpretation of the statute is reasonable. That is the only question on which we must rule. The order of the District Court, finding that both vested and non-vested terminees must be considered in its partial termination analysis, is affirmed.3

We next turn to the question of whether terminations occurring in multiple plan years can be combined in determining partial termination. Initially, we note that this issue has been addressed by only two other courts (excluding the District Court below) and both have found that it is permissible to aggregate terminations for multiple years. Weil, 750 F.2d at 13, n.2; Gulf Pension Litigation, 764 F. Supp. at 1167-68.4 We add our voice to theirs.

Matz argues that it is proper to combine all participant terminations in 1994, 1995 and 1996 because Household engaged in a corporate reorganization during that time.5 The Plan counters by arguing that because this a tax event, the relevant period should be one year, as are tax years. It also argues that "[a]ll of the published Revenue Rulings dealing with the issue of partial terminations" examined only terminations occurring within a single plan year. We note, however, that in all of the decisions cited by the Plan, the event that affected the plan participants occurred within one year. Thus, those cases are inapposite and unpersuasive in our analysis.

Neither the statute nor its legislative history specify whether aggregation is permissible. The IRS has taken no position on the issue and because of that we receive no guidance from it. The only instruction we have comes from Treasury Regulation sec.1.411(d)-2(b)(1), which provides that we must consider "all of the facts and circumstances in a particular case." No framework is provided beyond that.

We hold that because there is nothing in the language of the rule itself that requires a significant corporate event to occur within a plan year, Matz can combine terminations from 1994, 1995 and 1996, provided that he show that the corporate events of those years were related. We believe this view reflects the realities of the modern corporate world. Mergers and corporate reorganizations have grown into large and complex events, see e.g., In re Gulf Pension Litigation, 764 F. Supp. 1149, and often cannot be completed in one year. Furthermore, to establish a rigid rule that only terminations in individual plan years can be counted allows an unscrupulous employer to terminate some participants in December of one year and January of the next year, thereby eviscerating both the purpose of protecting employee benefits and the purpose of prohibiting employers from reaping unfair tax benefits. We are convinced that the requirement that the multiple year terminations be proven related prevents a plaintiff from gaining undue advantage too.

For the foregoing reasons, the orders of the District Court, holding that both vested and non- vested participants both be counted and that multiple plan years may be aggregated, in considering whether the Plan was partially terminated, are affirmed.

AFFIRMED.